Models Model details

Qwen

Qwen: Qwen3.7 Max

qwen/qwen3.7-max

Qwen3.7-Max is the flagship model in Alibaba's Qwen3.7 series. It supports text input and output and is designed for agent-centric workloads, with particular strengths in coding, office and productivity tasks, and long-horizon autonomous execution. The model offers notable gains in coding and agentic performance over prior Qwen generations and supports explicit prompt caching for efficient repeated context use.

Input
$1.475

/ 1M tokens

Output
$4.425

/ 1M tokens

Context window
1M

tokens

Limits

Context

1,000,000

tokens

Max output

131,072

13% of the context window

Model facts

Author

Qwen

Tokenizer

Qwen

Modality

Text -> Text

Created

May 21, 2026

Publication status

Hugging Face: not mappedLicense: not publishedParameter count: not published

Modalities

Input · TextOutput · Text

Generation controls

Frequency penaltyInclude reasoningLogprobsMax tokensPresence penaltyReasoningResponse formatSeedStopStructured outputsTemperatureTool choiceToolsTop kTop logprobsTop p

Capability scores

22.5
Agentic index

Better than 54% of models compared

66
Coding index

Better than 69% of models compared

29.5
Intelligence index

Better than 71% of models compared

Agent tasks

Category performance

Godot games
Top 50%

1187 Elo

55.7% win

Web apps
Top 39%

1203 Elo

49.4% win

Android apps
Top 62%

1174 Elo

48.9% win

Full-stack apps
Top 50%

1190 Elo

47.7% win

Slide decks
Top 58%

1194 Elo

47.2% win

Agentic games
Top 79%

1161 Elo

46.6% win

Mobile apps
Top 65%

1148 Elo

44.9% win

HTML slides
Top 61%

1158 Elo

43.6% win

Creative tasks

Category performance

SVG
Top 31%

1215 Elo

55.4% win

3D
Top 27%

1267 Elo

54.6% win

Websites
Top 22%

1275 Elo

54.4% win

Code
Top 25%

1277 Elo

54.1% win

UI components
Top 32%

1270 Elo

53.4% win

ASCII art
Top 35%

1215 Elo

53.2% win

Games
Top 25%

1267 Elo

53.1% win

Data visualization
Top 22%

1276 Elo

52.9% win